Merge branch 'sdh-filter-na-fix' into 'main'

Handle NAs in filtering approved data function

Closes #91

See merge request water/computational-tools/surface-water-work/hyswap!75

hyswap/utils.py

@@ -26,22 +26,22 @@ def filter_approved_data(data, filter_column=None):
     .. doctest::
 
         >>> df = pd.DataFrame({
-        ...     'data': [1, 2, 3, 4],
-        ...     'approved': ['A', 'A', 'P', 'P']})
+        ...     'data': [1, 2, 3, 4, 5],
+        ...     'approved': ['A', 'A, e', 'A', 'P', 'P']})
         >>> df.shape
-        (4, 2)
+        (5, 2)
 
     Then filter the data to only return approved data.
 
     .. doctest::
 
         >>> df = utils.filter_approved_data(df, filter_column='approved')
         >>> df.shape
-        (2, 2)
+        (3, 2)
     """
     if filter_column is None:
         raise ValueError("filter_column must be specified.")
-    return data[data[filter_column].str.contains("A")]
+    return data[data[filter_column].str.contains("A", na=False)]
 
 
 def rolling_average(df, data_column_name, data_type,

tests/test_utils.py

@@ -8,11 +8,11 @@
 class TestDataFilters:
     def test_filter_approved_data(self):
         """Test the filter_approved_data function."""
-        data = {"a": ["A", "A, e", "A, R", "A, [4]", "P", "P"],
-                "b": [1, 2, 3, 4, 5, 6]}
+        data = {"a": ["A", "A, e", "A, R", np.nan, "A, [4]", "P", "P"],
+                "b": [1, 2, 3, np.nan, 5, 6, 7]}
         data_df = pd.DataFrame(data)
         df = utils.filter_approved_data(data_df, filter_column="a")
-        assert df["b"].tolist() == [1, 2, 3, 4]
+        assert df["b"].tolist() == [1, 2, 3, 5]
 
     def test_filter_approved_data_error(self):
         """Test the filter_approved_data function."""